Michael Leon Chrzan

Data Scientist bridging K12 education practice, policy, and advanced analytics through machine learning, causal inference, and NLP to drive educational equity and evidence-based improvement at scale.

Beliefs Shape Reality - Parental Equity Beliefs Study

Structural equation modeling examining the relationship between parental equity beliefs and academic achievement. This research explores how parents’ beliefs about educational equity influence student outcomes and investigates potential mechanisms through which these beliefs shape children’s educational experiences.

San Francisco School Closure Scenario Analysis

Built a scenario generation algorithm evaluating 400+ viable school closure scenarios for San Francisco Unified School District. Developed equity-focused school quality indicators uncorrelated with historical inequities to help the district make data-informed decisions that prioritize student outcomes and minimize harm to vulnerable communities.

Gentrification & Educational Outcomes in Detroit

Modified the Pearman & Greene (2022) model to examine city-level impacts of gentrification on educational outcomes in Detroit. This analysis explores how neighborhood change affects school performance, student mobility, and educational opportunities in urban contexts.

Duolingo User Persistence Prediction

Semi-supervised learning approach to predict user retention based on engagement patterns in language learning. Analyzed user behavior data to identify factors associated with long-term persistence and provide insights for improving user engagement and learning outcomes.